SCIENCE NEEDS AND NEW TECHNOLOGY FOR INCREASING SOIL CARBON SEQUESTRATION

F. BLAINE METTING 1, JEFFREY L. SMITH 2, JEFFREY S. AMTHOR 3 and R. CESAR IZAURRALDE 4

1 Pacic Northwest National Laboratory, Richland, WA 99352, U.S.A. 2 U.S.D.A. Agricultural Research Science, Pullman, WA 99164-6421, U.S.A. 3 Oak Ridge National Laboratory, P.O. Box 2008, Oak Ridge, TN 37849, U.S.A. 4 Pacic Northwest National Laboratory, Washington, D.C. 20004, U.S.A.

Abstract. Fossil fuel use and land use change that began over 200 years ago are driving the rapid increase in atmospheric content of CO2 and other greenhouse gases that may be impacting climatic change (Houghton et al., 1996). Enhanced terrestrial uptake of CO2 over the next 50 to 100 years has been suggested as a way to reclaim the 150 or more Pg carbon (C) lost to the atmosphere from vegetation and soil since 1850 as a consequence of land use change (Batjes, 1999; Lal et al., 1998a; Houghton, 1995), thus effectively buying time for the development and implementation of new longer term technical solutions, such as C-free fuels. The ultimate potential for terrestrial C sequestration is not known, however, because we lack adequate understanding of (1) the biogeochemical mechanisms responsible for C uxes and storage potential on the molecular, landscape, regional, and global scales, and (2) the complex genetic and physiological processes controlling key biological and ecological phenomena. Specically, the structure and dynamics of the belowground component of terrestrial carbon pools, which accounts for two-thirds of global terrestrial organic C stocks, is poorly understood. Focusing primarily on forests, croplands and grasslands, the purpose of this chapter is to consider innovative technology for enhancing C sequestration in terrestrial ecosystems and address the scientic issues related to better understanding of soil C sequestration potential through appropriate and effective approaches to ecosystem management.

1. Soil Carbon Sequestration Potential 1.1. NATIVE ECOSYSTEM POTENTIAL Land use options for enhanced C sequestration at the landscape and regional scales include protection and selective management of native ecosystems, and use of appropriate and advanced management practices in manipulated ecosystems. Of course, there is not always a clear distinction between these two approaches, with constructed wetlands, managed grazing of rangelands, and multiple use national forests being some examples. As a baseline for judging C sequestration potential, an understanding of inherent ecosystem productivity is important. The ratio of net plant growth (gross production minus respiration) to the amount of absorbed photosynthetically active radiation (PAR) is known as the light (or radiation) use efciency ().1 In
Climatic Change 51: 1134, 2001. 2001 Kluwer Academic Publishers. Printed in the Netherlands.

12

F. BLAINE METTING ET AL.

Table I Light use efciencies among ecosystem types (Ruimy et al., 1994) Ecosystem type Equatorial moist forest Equatorial evergreen tree plantation Mediterranean evergreen forest Temperate deciduous forest Temperate deciduous tree plantation Temperate and subpolar coniferous forest Temperate and subpolar coniferous tree plantations Temperate grassland C-3 crops C-4 crops All cultivated/managed vegetation Mean (g/MJ)a 0.62 1.74 0.37 1.01 2.72 1.57 1.69 1.26 2.71 3.51 2.07

a With respect to total (i.e., above- plus below-ground) plant growth.

an analysis of the literature, Ruimy et al. (1994) derived values for the light use efciency for a number of ecosystem types (Table I). The values range widely, being especially high for C-4 crops and low for some forest types. Establishing accurate values for is difcult, thus the estimates in Table I are only approximations. There is also considerable variation among individual studies of the same ecosystem type. Nonetheless, differences among ecosystem types are apparent, and these differences highlight the importance of photosynthesis and plant respiration in the C balance of the terrestrial biosphere and for understanding C sequestration potential. In particular, the light use efciency of managed ecosystems generally exceeds that of unmanaged native ecosystems. Whereas light use efciency need not be strongly related to long-term C sequestration, it does represent the input side of the C balance equation for a given light environment and canopy structure. Differences in photosynthetic capacities among ecosystems contribute to the importance of land cover or land use change on potential C sequestration. Instantaneous photosynthetic responses of plant canopies to light differ among ecosystem types. Based on a review and analysis of micrometeorological and canopy-enclosure measurement data, Ruimy et al. (1994) concluded that photosynthetic capacity per unit ground area (i.e., CO2 uptake at high irradiation, 1800 mol photons PAR/m2 sec) is greatest in crops, followed by grasslands and forests (Table II). The lower apparent photosynthetic capacity in C-4 grasslands compared to C-3 grasslands, in spite of the generally greater leaf-level photosynthetic capacity in C-4 species, may be attributable either to the limited data from grasslands or differing degrees of stress for the measured C-3 and C-4 grasslands.

INCREASING SOIL CARBON SEQUESTRATION

13

Table II Instantaneous photosynthetic capacity (CO2 uptake at PPFD a = 1800 mol/m2 /s) of plant canopies in different ecosystems (Ruimy et al., 1994) Ecosystem Photosynthetic capacity (mol CO2 /m2 /s) 21 19 31 23 27 33 20 24 28

Broadleaf forests Conifer forests C-3 b grasslands C-4 grasslands C-3 crops C-4 crops All forests All grasslands All crops

a PPFD = photosynthetic photon ux area density. Midday, clear sky summer PPFD is typically 18002100 mol/m2 /s at middle latitudes. b Plants relying exclusively on the classical Calvin cycle x CO2 as the C3 compound phosphoglyerate are termed C-3. In contrast, C4 dicarboxylic acids are the initial products of CO2 -xation in C-4 plants (Sage and Monson, 1998).

Table III includes estimates of the potential for sustained terrestrial (soil + vegetation) C sequestration in native and managed ecosystems developed by a group of scientists sponsored by the Department of Energy, and although the data and tables are from non-peer reviewed sources, the numbers used are documented (DOE, 1999). In descending order, the relative potential for C gain in native ecosystems in the next few decades is probably greatest in tropical savannah, followed by tropical forests, wetlands, and unmanaged temperate grasslands and forests (approximately equivalent). With hypothesized global warming and associated accelerated soil organic matter (SOM) decomposition, the potential for signicant loss of soil C is thought to be greatest in high latitude ecoregions, including peat lands and arctic and boreal tundra and taiga (Amthor and Huston, 1998). 1.2. MANAGEMENT OF CROPLANDS , FORESTS AND GRASSLANDS On a global scale, it has been suggested that soils may have a nite, steadystate C carrying capacity controlled by the interactive temperature and moisture components of climate on vegetation and by soil texture and mineral composition (Schlesinger, 1995). For a given soil in an undisturbed or minimally disturbed

14

F. BLAINE METTING ET AL.

Table III Sustained terrestrial C sequestration potential. The primary C sequestration method is rated with High (H), Medium (M), and Low (L) levels of sustained management intensity required over the long term. Global potential C sequestration (CS) rates were estimated that might be sustained over a period of up to 50 years (DOE, 1999)a
Ecosystem Agricultural lands Biomass crop lands Grasslands Rangelands Forests Wetlands Urban forest and grass lands Deserts and degraded lands Sediments and aquatic systems Tundra and taiga TOTAL Primary method to increase CS Management (H) Manipulation (H) Management (M) Management (M) Management (M) Restoration, creation and maintenance (M) Creation and maintenance (M) Manipulation (H) Protection (L) Protection (L) Potential CS (GtC/y)b 0.850.90 c 0.50.8 d 0.5 e 1.2 f 12 g 0.10.2 h i 0.81.3 j 0.61.5 k 0.10.3 5.658.71

a DOE (1999). Chapter 4. Carbon sequestration in terrestrial ecosystems. b R&D allows improvements in carbon sequestration implementation. Management intensity includes fer-

tilization, irrigation, pesticides, and heavy equipment usage. No reallocation of land use, except for 1015% of agricultural land to biomass crop lands. Baseline values are from Amthor and Huston (1998). c Soil carbon only; recovery of an amount equivalent to what was lost from native soils prior to agricultural use; implementation of best-available management (e.g., no-till, intensied production and residue inputs, intensied rotations with crop rotation, double cropping, greater use of perennials) and new technologies, such as precision farming. d An average annual aboveground productivity level of 13.2 Mg/ha/y. Belowground C storage is 1.75 Mg/ha/y and assumed to be permanent and to not provide any negative feedback on further storage; Short rotation woody crop and perennial grass production is assumed to provide equivalent C storage benets; The energetic costs of producing and harvesting switchgrass results in a biomass energy return ratio (energy in harvested biomass divided by production energy costs) of 12.3 and an energy gain of 343% for ethanol production. The C gain from substitution of ethanol for gasoline (2.48 MgC/ha/y) after subtracting carbon costs of production (0.60 MgC/ha/y) and adding an average belowground sequestration rate of 1.75 MgC/ha/y provides annual C savings of (2.48 + 1.75 0.60) = 3.60 MgC/ha/y; trees and grasses are assumed to be equally efcient at net C production and sequestration and that production of ethanol and electricity provide equivalent net benets in terms of C savings; a conversion of 10% of current cropland to biomass crops for energy represents a realistic target, while under more favorable conditions a 15% conversion might be achievable on a world basis. e Intensication of management with fertilization, controlled grazing, and species improvements; 25% increase in belowground carbon stocks; linear increases through 2050. f Total increase of 27 GtC through 2050; rehabilitation of degraded range land and fertilization by increasing CO2 . g Watson et al. (1996) estimate 11.6 GtC/y (their Table 14) and include above- and belowground vegetation, soil C, and litter. Their estimate does not include R&D to increase carbon sequestration. Trexler (1998) suggests a rate a 2 GtC/y may be plausible. With focused R&D, both these values may be exceeded. h The wetlands estimate is from Armentano and Menges (1986). It is for temperate and boreal wetlands, and it primarily represents C lost to disturbance of wetlands by agriculture, forestry and peat harvesting. The range for the tundra was taken from Oechel et al. (1993) and represents the impact of regional warming on net C balance. Thus, the two estimates may overlap in global coverage, but the rst represents direct anthropogenic disturbance that is presumably reversible, and the second represents an indirect impact that would be difcult to manage. Tropical wetlands are not included in the estimate. i No estimate available. j From Table 23 of Lal et al. (1999). Soil C emphasis; erosion, desertication and global warming effects are controlled; includes restoration of lands, reclamation of salt-affected soils, agricultural intensication on non-degraded lands (0.015 GtC/y), and fossil fuel C offset of 0.2 GtCg/y; includes accretion of inorganic carbonates. k Estimate from Stallard (1998).

INCREASING SOIL CARBON SEQUESTRATION

15

ecosystem, however, the maximum carbon sequestration potential is not known in part because soils take more time to reach equilibrium than vegetation. For managed ecosystems, it may be possible to increase the soil carrying capacity for C through plant species selection or by altering the microclimate via nutrient and water management and other means. The principal approaches for increasing terrestrial C sequestration are converting marginal land to more productive grasslands and forest, increasing productivity on crop and forest land with residue management to slow organic matter decomposition, management approaches to reduce C loss and the application of technology. On a global basis it is difcult to estimate the potential for increasing C sequestration because baseline inventory data are inadequate; the difculty is compounded by varying and uncertain use of terms such as marginal cropland, grassland and degraded soils. We present here an analysis of the potential for C sequestration on U.S. soils for which there are adequate data on land use and soil C inventories (Table IV). Based on such information, conclusions regarding C sequestration potential of managed systems should be applicable wherever in the world local land use and economic conditions are known. Because management options for increasing C already exist, forests and croplands can be usefully evaluated to address the consequences of changing land uses. Thus, the discussion centers on manipulation of forests, agricultural land and, to a lesser extent, grazed and ungrazed grasslands. There are 47 million hectares (Mha) of marginal cropland in the United States, 14 Mha of which have been converted to grassland under the Conservation Reserve Program (CRP). Thirty-three Mha remain available for conversion or restoration to high C sequestrating vegetation, with the only constraint being guaranteed profitability to the landowner. Current CRP land is concentrated in the mid-U.S. from Texas to Minnesota.2 While the principal CRP benet is reduced soil erosion, ancillary benets, such as enhanced wildlife habitat, improved air quality, and improved surface water quality make the estimated benet of this program $3.4 to $11 billion annually (Young and Osborn, 1990). The C sequestration potential of CRP land has been estimated to range between 0.4 and 1.0 metric tons (MT) C/ha/yr, for a total annual sequestration potential of between 6 and 14 106 MT C (0.0060.014 Pg C) (Lal et al., 1998b). It has been estimated that about 25 of the 47 Mha of degraded cropland and pastureland in the U.S. are suitable for growth of softwood tree species and 22 Mha for hardwoods (Parks, 1992). Some of this land is taken up in the CRP so that the following applies to the remaining 33 Mha. Approximately 10 Mha were used for an analysis of softwoods because economic data were available to indicate this conversion would maintain a 4% annual return on investment. Considered a low estimate, the analysis suggested an annual C storage increase similar to CRP land of 0.033 Pg C (0.57 MT C/ha). As shown in Table IV, if all available land in the United States was converted with similar production rates (biological potential) the annual C sequestration would be 0.119 Pg C.

16

F. BLAINE METTING ET AL.

Table IV Annual U.S. potential for C sequestration from managed forests, arable lands and pastures Strategy Average C sequestration Low estimate High estimate Pg C/year Pg C/year

Forestry a Converting marginal crop/pasture to forest Increasing timber growth on timber land Growing short-rotation woody crops for energy Increasing tree numbers/canopy cover in urban areas Planting trees in shelter belts Subtotal Arable land b Cropland conversion to CRP (excluding agroforestry) Soil restoration (eroded land, mine land, salt affected soil) Conservation tillage/residue management Better cropping systems (fertilizer, cover crops, manure) Subtotal Total managed forests, arable land, pastures
a After Hair et al., 1996; Birdsey et al., 1992. b Lal et al., 1998b.

0.033 0.138 0.091 0.011 0.003 0.276

0.119 0.190 0.180 0.034 0.006 0.529

0.006 0.011 0.035 0.024 0.075 0.351

0.014 0.025 0.107 0.063 0.208 0.737

The second overall approach for enhancing C sequestration is to increase the productivity of crop and forest land. In agriculture it has been shown that high yields can be maintained while simultaneously managing land for optimum C storage in the form of increased surface residue and SOM. In this respect the adoption of conservation tillage practices is a viable mechanism to increase C storage by (1) reducing erosion, (2) increasing soil aggregation, and (3) decreasing the loss of SOM to microbial oxidation that results from tillage (Lal et al., 1998b). However, the adoption of conservation tillage has been somewhat slow because yields can be depressed in the rst few years and pesticide and herbicide use often increases. Research is nding ways to overcome these problems and, as of 1997, 37% of all U.S. cropland was under some form of conservation tillage. The C sequestration potential of the combined practices of no-till, mulch, and ridge tillage was estimated to be 14.1 106 MT (0.014 Pg) C/yr, with associated savings in fossil fuel equivalent to 1.6 106 MT C/yr. In addition, managing crop residues from these systems may sequester another 22.5 106 MT (0.023 Pg) C/yr (Cole et al., 1996).

INCREASING SOIL CARBON SEQUESTRATION

17

The U.S. Forest Service has estimated that 85 Mha of forest land have the biological potential to increase production through regeneration and stocking control. Vasievich and Alig (1996) showed, for an economic constraint of a 4% annual return on investment, that these timberlands could sequester 0.138 Pg C/yr under proper management. They also estimated that if all timberlands in this analysis, regardless of economic constraints, were managed for C sequestration, the potential would increase to 0.19 Pg C/yr (Table IV). Another strategy for sequestrating C is the conversion or reallocation of agricultural land to woody crops. The U.S. Department of Agriculture (USDA, 1990) projected that 52 Mha of agricultural land will be in excess of demand for food production from 2000 to 2030. Wright et al. (1992) estimated that between 14 Mha and 28 Mha of cropland are suitable for woody crops. With a current production level of 6.5 MT C/ha, the sequestration potential for this strategy is 0.09 to 0.18 Pg C/yr (Table IV). Other sequestration potential for both forestry and agriculture is also included in Table IV. Lal et al. (1998b) estimated a total potential for agriculture between 0.075 and 0.208 Pg C/yr. The totals for forestry and agriculture range between 315 and 633 106 MT C/yr (mean of 0.47 Pg/yr). The lower and upper total values in Table IV represent 22% to 44% of the total U.S. fossil C emissions of 1.44 Pg, as estimated by the U.S. Department of Energy, and slightly less when using the total reported by the U.S. Environmental Protection Agency (1.709 Pg). Grasslands constitute approximately 26% of the total U.S. land area, equivalent to 240 Mha, in comparison to cropland (134 Mha) and forests (298 Mha). On a worldwide basis temperate grasslands occupy 900 Mha. Grasslands are among the most productive systems in the world with annual NPP of 400500 g C/m2 compared to savanna (300350 g C/m2 /yr), temperate forests (550-600 g C/m2 /yr) and tropical forests (8001000 g C/m2 /yr) (Burke et al., 1997). Native grasslands in North America are designated mixed grass, short grass, or tall grass prairie. Coupland (1992) estimated net above-ground production for mixed grass prairie to range from 250 to 600 g/m2 /yr, with below-ground production ranging from 300 to 1000 g/m2 /yr. For short grass prairie, the aboveground NPP was estimated at 50 to 325 g/m2 /yr, and belowground between 540 to 790 g/m2 /yr (Lauenroth and Milchunas, 1992). Kucera (1992) collated information on tall grass prairie and reported a range of 200 to 1000 g/m2 /yr, with belowground production being at least equal to aboveground biomass production. Worldwide, temperate grasslands have similar productivity ranges. To increase production, it is necessary to bring these lands under management with the option of choice being N fertilization. This could signicantly increase NPP. However, grasslands have an inherent capacity to emit N2 O, a strong greenhouse gas (Mummey et al., 1999), and this would likely increase with the use of N fertilizer. Thus, a comprehensive analysis that includes NOx emissions and the C cost of fertilizer production and application is needed to evaluate the net C sequestration benets of fertilization of grasslands.

18

F. BLAINE METTING ET AL.

Degraded soils also represent a large potential for C sequestration. Worldwide, there are approximately 1965 106 ha of degraded soils, 4% from physical degradation, 56% from water erosion, 28% from wind erosion, and 12% from chemical degradation (Oldeman et al., 1991). With proper management these soils have the combined potential to sequester between 0.81 and 1.03 Pg C/yr. The best options include reclamation of saline soils, erosion control, restoration of eroded lands, and biofuel production (Lal et al., 1999).

2. New Technology for Soil Carbon Sequestration New technology with the potential to enhance soil C sequestration can be categorized as (1) technology for soil, crop and forest management, (2) exploitation of underutilized land resources and existing biodiversity, (3) plant biotechnology, (4) microbial biotechnology, and (5) chemical technology. The U.S. Department of Energy (1997) made theoretical considerations on the potential of new soil C sequestration technologies, their probability of success and the time required for their implementation. These considerations are reproduced and further expanded in Table V. The likelihood of success refers only to the technology being achievable; it does not consider economics or public acceptability. None of the entries in Table V consider additive potential from precision management; nor does any category include additive input from any other, although they probably overlap. 2.1. PRECISION TECHNOLOGY FOR CROP AND FOREST MANAGEMENT As discussed above, selected management practices in agriculture and forestry have the potential to signicantly enhance soil C sequestration. In addition, sequestration potential could be further enhanced with new technology. Ground and plant based sensing mechanisms linked to proximal or remote imaging with computer control comprises the emerging set of technologies known as precision agriculture (Pierce and Sadler, 1997). In time, the widespread use of precision methods over broad areas of agriculture and silviculture holds tremendous potential for use in efforts to enhance soil C sequestration or, minimally, indirectly inuence atmospheric CO2 increase though reduced energy consumption. The contribution of this high technology to precision farming and forestry will include the following: Sensors to detect the appearance of pathogens for early, precisely targeted control. Use of aerial and global positioning technology for precise application of fertilizer how much, when and where it is needed in a eld or forest. Efcient just in time irrigation systems that maximize water use efciency. Each of these approaches will apply rapidly evolving computer and information technology. In the near term, sensors will be electronic devices. In the longer term,

INCREASING SOIL CARBON SEQUESTRATION

19

Table V Proposed global carbon sequestration potential from new technology Technical approach Potential sequestration (Pg C/yr) Potential success High/Med./Low (inverse of risk) Time to implement (yr)

Crop and forest management a Cropland management Forest management Under-utilized resources Deserts and saline/alkaline aquifers b Native plant biodiversity c Plant biotechnology d Innate productivity Photoassimilate partitioning Stress tolerance (salt, drought) Nitrogen xation Lignin (ber) biosynthesis Bio products e Microbial biotechnology Soil conditioners Mycorrhizal fungi Engineering rhizobial communities Chemical technology Smart fertilizers Soil additives/conditioners Calcium for arid regions Plant growth regulators 01 00.5 02 01 M H L H 10 Now Now Now 00.5 12 12 M H H Now 10 25 0.52 13 H M/H Now Now

01 N/A

L H

25 Now

16 02 13 12 12 N/A

L/M L/M M L/M L/M H

25 25 10 25 5 5

a Does not include impacts of biotechnology or chemical technology, which are probably additive. b Baseline is unmanaged arid/hyper-arid desert without plant cover. Lower value is periodic desert ooding. Larger value is for highly engineered microalgal mass culture. c Because many scenarios are possible, there is no single comparison base. d All of the plant biotechnology options are intimately linked, so values are probably not additive. e Fossil fuel offsets.

20

F. BLAINE METTING ET AL.

plants may be genetically engineered to produce unique spectral signals in response to stress such as pathogen attack or water decit, for detection by novel and costeffective remote sensing systems (e.g., hyperspectral imaging) now in the early stages of development. 2.2. UNDERUTILIZED RESOURCES Underutilized resources include land, groundwater, and native biodiversity. For land and groundwater, the related cases considered here are (1) to irrigate deserts with water from saline and alkaline aquifers, or (2) to use the saline/alkaline groundwater to mass culture microalgae. Microalgae, including cyanobacteria (blue-green algae), are natural components of biological crusts in semiarid lands and deserts and are capable of xing C and N at high temperature and irradiance when provided with sufcient moisture. Although never attempted, it is theoretically possible to use periodic ooding with saline/alkaline groundwater to encourage the growth of salt and alkaline-tolerant cyanobacterial inoculants over large areas of desert (Knutsen and Metting, 1991). In addition to photosynthesis, enhanced C sequestration through carbonate formation might also take place if the groundwater were rich in Ca.3 Use of the same groundwater resources to mass culture microalgae in highly engineered systems in deserts has been the subject of ongoing research for over 25 years. Although microalgae are cultivated commercially for high value products (Metting, 1996), the large-scale production of microalgae strictly for biomass or C capture has yet to be attempted because of unfavorable economics. Because the single largest projected cost is the acquisition and provision of CO2 , it may be feasible to transport sequestered C from power plants to deserts for this purpose. The more practical near-term opportunity is to exploit underutilized native plant resources with, for example, inherent tolerance to acid soil, salinity, drought or other stresses for purposes such as reclamation of degraded land and erosion control to stem desertication. This might be accomplished with a focus on selected C-3, C-4, and CAM4 plants for water use efciency and by expanding the area planted to shelter belts and to food and biomass crops shown to be effective under various ecological situations. Lal et al. (1999) consider in detail the use of native plant biodiversity for restoring degraded lands, the effects of which on soil C sequestration are not well understood. 2.3. PLANT BIOTECHNOLOGY Plant genetic engineering for enhanced NPP could lead directly to greater C sequestration. Changes in traits other than NPP, such as chemical composition or photosynthate partitioning between shoot and root, could also inuence C sequestration. Some target endpoints for enhanced C sequestration via plant biotechnology are to:

INCREASING SOIL CARBON SEQUESTRATION

21

Improve innate photosynthetic efciency and net primary production (NPP). Manipulate photoassimilate partitioning. Manipulate content of lignin and other polymers. Develop various bio-products to displace fossil fuel based products. Improve stress tolerance, including salt and drought tolerance, and Al tolerance for acidic tropical soils. Engineer C-4 photosynthesis into C-3 plants. Engineer N2 -xation into non-leguminous plants. Two general points emerge from such a list. First, the application of most or all of the technologies would likely be limited to highly managed ecosystems, such as croplands, managed forests and biomass plantations. In addition to signicant logistical challenges, the introduction of genetically altered characteristics into native plant communities and ecosystems could engender societal resistance based on environmental and ethical concerns. Second, many of the proposed modications have been the subject of research for decades, with limited success. Notable exceptions include altered photoassimilate partitioning in favor of seeds in grain crops and directed reductions (rather than increases) in lignin content of silage. Together, these experiences imply that large-scale changes beyond agriculture and silviculture, and the chances for large changes in key plant characteristics are limited. This may largely be because many of the proposed changes to plant processes involve multiple genes and the coordination of biochemistry with anatomy, as well as complex and interacting metabolic pathways (ap Rees, 1995). This suggests that directed genetic solutions to complex traits will be difcult, much more so than changes involving only one (or a few) genes. Against that backdrop, we appraise each point briey. Innate photosynthesis and NPP might be improved in C-3 species if the characteristics of ribulose 1,5-bisphosphate carboxylase/oxygenase (rubisco) could be modied. An obvious target is the CO2 /O2 specicity of this bifunctional enzyme. If the specicity of rubisco for CO2 could be improved, net photosynthesis would increase by reducing or eliminating photorespiration. A simple measure of the potential for increased photosynthesis is the known ratio of photorespiration to photosynthesis. With present day atmospheric CO2 and O2 concentrations, the photorespiration/photosynthesis ratio varies from 0.1 to 0.3 for most C-3 plants (Amthor, 1995). Theoretically, photosynthesis could thus be increased 10 to 30% if photorespiration were to be eliminated by engineering rubisco. Although molecular approaches to modifying plant physiology are advancing, decades of previous research have contributed little improvement to photosynthesis per se, except as associated with canopy architecture and plant nutrition. Of course, increased photosynthetic capacity would have to be coupled with increased sink activity to obtain greater NPP. For purposes of soil C sequestration, a key aspect of partitioning is the relative allocation to roots. Thus, modication and control of root architecture and rooting

22

F. BLAINE METTING ET AL.

depth is an important target whose realization depends on improved understanding of underlying genetic and metabolic processes (Schiefelbein et al., 1997). Roots play a fundamental role in C cycling and SOM stabilization but our quantitative understanding of their growth and turnover rates is rather incomplete. The study of roots is difcult because they grow in a porous medium in close interaction with minerals and other living organisms. The use of C isotopes in root studies, however, has steadily enhanced our understanding of their role in SOC dynamics. For example, Balesdent and Balabane (1996) assessed the extent to which maize roots contributed to SOC content by using natural 13 C techniques. Their isotopic data revealed that roots had contributed 58% more C to SOC content than that supplied by leaves and stalks together. Crop breeders have increased the harvest index (ratio of grain to total aboveground biomass) in cereals, but the accompanying changes in the ratio of aboveground to below-ground growth have varied (Evans, 1993). If increased root growth comes at the expense of grain or other harvest targets, farmers would likely be hesitant to adopt cultivars with these characteristics. Information on root growth characteristics could be an important aspect to consider when selecting plant cultivars. For example, Xu and Juma (1993) used 14 C pulse labeling to discern root growth characteristics and C stabilization in two barley cultivars. Although Samson, a six-row semi-dwarf feed cultivar, produced less root mass than Abee, a two-row medium-height feed cultivar, it stabilized more C in soil than Abee. Swinnen et al. (1995) also used 14 C pulse labeling to study rhizodeposition of winter wheat and spring barley grown with conventional and integrated farming methods. Soil management did not affect root growth in wheat but it did affect that of barley. Unexpectedly, conventional farming methods led to greater root growth, root respiration, and rhizodepositional uxes than those measured under integrated farming methods. The amount of C contributed annually by rhizodeposition was twice as much as that contributed by the standing roots left at harvest. Further research in this area is needed to improve our understanding of the role that roots play at the local and global scales (Jackson et al., 1997). Modication of the lignin content and structure of plants is another potential biotechnology for enhancing C sequestration. Vascular plants may contain up to 20% of their dry weight in lignied C compounds. These highly aromatic materials provide mechanical support and defend against pathogen attack. The insolubility and complexity of lignin polymers render them resistant to degradation by most microorganisms and more persistent in soils than cellulose and other non-aromatic compounds. There are several ongoing efforts to modify lignin content which primarily seek to reduce content for increased ruminant digestibility and better paper pulp (Sewalt et al., 1997). The basic knowledge gained from these efforts to control the synthetic pathways could perhaps be directed to the opposite purpose of increasing lignin production, for example, in the roots of select grasses. It is important to realize that the quality of lignin varies among plant species. Also, fungi and other microorganisms have evolved to use lignin as a C source and

INCREASING SOIL CARBON SEQUESTRATION

23

microbial communities can change so that enhanced microbial degradation may offset increased lignin input to soil. Any genetically based improvements in productivity of biofuel crops have the potential to reduce fossil fuel use. While this is not a direct sequestration strategy, any reduction in fossil fuel use reduces the total requirement for C sequestration. In this case, it is in addition because biofuel production is sustainable and does not result in ecosystem C losses that are greater than the fossil fuel savings. Improved stress tolerance would result in greater NPP and could also increase the distribution of plant species. Key stresses relevant to enhancing C sequestration include drought and salinity, which affect vast areas of arid and semi-arid ecosystems; Al stress in acidic tropical soils is also of considerable importance over large areas. As with other complex ecological interactions, stress tolerance is generally determined by multiple biochemical pathways and signaling networks that control acquisition of water and nutrients, chloroplast function, and synthesis of stress proteins and osmotically-active metabolites. To date, the transfer of individual genes has resulted in only marginal enhancement of stress tolerance demonstrating again the necessity for understanding multiple gene interactions (Bohnert and Jensen, 1996). The benets in terms of enhanced NPP and potential C sequestration that could be associated with improved stress tolerance justify a concerted research effort in this area. Net primary production by plants with C-4 photosynthetic metabolism, most notably tropical grasses, can exceed that of C-3 plants, at least in environments to which they are adapted. Therefore, replacement of C-3 with C-4 physiology in select species might be an avenue toward enhanced C sequestration in terrestrial ecosystems. Importantly, even though C-4 plants account for no more than 1% of the number of higher-plant species, they may already account for 2025% of global photosynthesis in non-crop ecosystems (Lloyd and Farquhar, 1994); so C-4 photosynthesis is already active in the terrestrial C cycle. The notion of C-4 wheat (by nature a C-3 species) has been of interest to crop scientists for some time. However, C-4 photosynthesis is associated with an integrated complex of biochemical and anatomical features that are not easily introduced into C-3 plants, even though C-4 photosynthesis/anatomy has apparently evolved on several separate occasions (Furbank and Taylor, 1995). Photosynthesis in C-4 plants depends on unique leaf anatomy, so that inserting C-4 photosynthesis into C-3 plants using biotechnology has not yet been achieved, although a recent study reported successful expression of a C-4 enzyme in rice, a C-3 plant (Ku et al., 1998). Also, C-4 photosynthesis can be inferior to C-3 photosynthesis in cold environments, so the utility of C-4 photosynthesis is geographically limited. Moreover, woody plants, which almost universally lack C-4 metabolism, may be most useful for storing C over time periods of a few decades. This too indicates a limited prospect for sequestering large amounts of C globally by engineering a shift from C-3 to C-4 photosynthesis unless it could be inserted into forests, which

24

F. BLAINE METTING ET AL.

seems unlikely outside the scope of tree plantations, if it could be introduced into trees in the rst place. Another long-sought approach to greatly improving the efciency and reducing the cost of crop production is by engineering N2 -xation into plant species (particularly grasses) other than legumes and the few other plant families that harbor appropriate bacterial symbionts. The minimal progress toward introducing symbiotic N2 -xation into non-legumes again probably reects an inherent limitation for directed modication due to the complex, multigenic nature of symbiosis in both the plant and microbial partners. 2.4. MICROBIAL BIOTECHNOLOGY Microbiological approaches to enhanced C sequestration are also possible. As with plants, near-term opportunities include screening and selection of existing biodiversity to improve microbial-plant symbioses, such as mycorrhizal fungi, bacterial N2 -xation, biological control, phosphate solubilization, and soil conditioning (Metting, 1992). Developing the capability to grow mycorrhizal fungi (and other unculturable microorganisms) in pure culture would be tremendously valuable. These fungi are symbiotic with all important forest tree species and most crops, for which they enhance water gathering efciency and nutrient (P and microelements) uptake. Genetic engineering to improve these traits, as well as enhancing the energy efciency of N2 -xation in free-living and symbiotic bacteria, is another mid- to long-term (1025 yr.) research target. Research has begun with coupled plant-microbial systems with the aim of manipulating entire microbial rhizosphere communities by modulating root exudate patterns (Savka and Farrand, 1997). In order to achieve these objectives, it will rst be necessary to better understand the intricacies of symbiosis and the very complex ecology of the rhizosphere. Finally, microbial production of polysaccharides and humic materials facilitates formation and stabilization of aggregates which, in turn, provides a measure of physical protection from degradation. Microbial inoculants for conditioning soil have been developed and used on a limited scale (Metting, 1992, 1996). 2.5. CHEMICAL TECHNOLOGY Other potential new technology for enhanced soil C sequestration includes novel fertilizers and soil amendments based on advances in materials research, and the use of plant growth regulators (PGRs). Smart fertilizers might be developed to release nutrients upon demand by the plant in response to specic molecular signals. Chemical products (e.g., polyvinyl alcohol, lignites) to promote and stabilize soil aggregation have been introduced for erosion control in the past, but have not been economical for large-scale use in agriculture. For carbonates to act as a sink for atmospheric CO2 , a source of Ca is required. A net transfer of C to the soil does not take place if the Ca is derived from dissolved carbonate to begin with. This is particularly important for alkaline soils in arid

INCREASING SOIL CARBON SEQUESTRATION

25

and semi-arid regions where increased irrigation or precipitation without Ca input could result in net emissions of CO2 from carbonate dissolution (Grossman et al., 1995). If economical sources of Ca are identied, the potential exists to enhance carbonate formation in alkaline soils by means of Ca addition. Microbial processes for carbonate formation may also exist in soil similar to those in aquatic and marine systems, but these have not been extensively studied. Plant growth regulators (PGRs) are synthetic and natural compounds (mostly organic), including hormones, that affect plant metabolism and development when present in very low concentrations. Various PGRs have been used commercially for many years to hasten or delay the ripening of fruit, to induce senescence, increase production of a desired metabolite (e.g., latex, oleoresins), or control plant morphology, including tillering in grasses. Molecular signals from symbiotic and pathogenic soil microorganisms are also known to inuence root architecture when present in low concentrations in the rhizosphere. Thus, the discovery, selection, and use of select PGRs to inuence belowground growth and metabolism may also be a feasible approach to enhance soil C sequestration.

3. Knowledge Gaps and Scientic Research Needs 3.1. THE OVERARCHING QUESTIONS The existence of many fundamental knowledge gaps highlights the need for scientic research to improve our understanding of soil C sequestration and establish an information base for the development and implementation of new sequestration technology. Ongoing and future research efforts and the identication of specic gaps in basic understanding with which they are associated can be addressed in the context of a primary set of overarching questions in need of resolution: How can we best reduce the large uncertainties in global terrestrial C inventories? Are estimated native ecosystem C sequestration capacities equivalent to their maximum carrying capacities? Is the historic (pre-agriculture) storage capacity of agricultural soils equivalent to their maximum inherent capacity? How accurate an estimate of the long-term potential for soil C sequestration can be made? Will tropical and north temperate forests, and high latitude taiga and tundra ecosystems become net sinks or sources of C with global warming? Will warming favor enhanced production or SOM oxidation? How will this vary among regions? How can other potential benets and risks of enhanced soil C sequestration, such as soil quality, best be identied and quantied? What is the potential for plant and microbial genetic engineering in the postgenome era to make signicant impacts on C sequestration? What are the

26

F. BLAINE METTING ET AL.

inherent biological and ecological limits to manipulating NPP, soil microbial communities, and plant-microbe symbioses? What are the opportunities for developing new biological, chemical, and C sequestration management technologies? Which of these can realistically be applied over large areas of the Earth? 3.2. SOIL CARBON INVENTORIES : DATABASE CONSISTENCY AND
VERIFICATION

More accurate baseline inventories of global land use, extent and sequestration capacity of native and disturbed ecosystems, and stocks of organic and inorganic soil C are needed. To be useful for long-term forecasting and planning, it is important not only to have more reliable present day inventories, but also of C stocks prior to and during expansion of agriculture in the nineteenth century. Current inventories for some regions, estimates of historic CO2 release to the atmosphere, and projected sizes of future terrestrial C sinks often carry uncertainties of 2550% or more. This results from inherent variability and heterogeneities across temporal and spatial scales, imperfect assumptions, and inadequate data collected mostly for other purposes. These data are statistically inadequate for representing large areas of the Earths land surface. For example, it has been speculated that the missing C sink may just be a result of misinterpretation of inadequate databases (Lal et al., 1998c). Baseline data and improved inventories that are statistically signicant and consistent over large areas of the Earth are needed. Specically, studies designed to collect information on the scale of kilometers or smaller are needed for large areas of Africa, South America, and Asia because existing data are not accurate enough for predictive and interpretive purposes. Also, special attention should be paid to categorization and prioritization of degraded lands for restoration and of saline, alkaline, and acid sulfate soils to enact programs to stop or reduce future degradation. Data collection should also be designed in such a way as to improve our ability to project inuences on soil C of land use changes such as, for example, urbanization, deforestation, afforestation and reforestation, or the creation of biofuel plantations. Information is also needed on a number of selected ecosystem parameters that are not now commonly collected in order to better understand soil C sequestration. In addition to aboveground biomass and SOC data, efforts should include data on litter and coarse woody debris, neither of which is routinely included in commercial forest inventories. Also, it is very important to quantify SIC, for which native stocks and dynamics and relationships to land use are virtually unknown. Example of collaborative efforts to improve databases is SOMNET, the International Soil Organic Matter Network (www.nmw.ac.uk/GCTEfocus3/networks/somnet.html) and the World Soil

INCREASING SOIL CARBON SEQUESTRATION

27

Resources of the Soil Conservation Service (WRS-SCS) effort to collate soil C data from around the globe (Eswaran et al., 1995). 3.3. SOIL CARBON DYNAMICS AND PEDOSPHERIC PROCESSES Despite recent progress toward improved national and regional soil C budgets, research and modeling of soil C dynamics, and research in land use and soil management, many knowledge gaps still remain in our understanding of the fundamental mechanisms responsible for soil C sequestration and interrelated pedospheric processes. For example, limited data are available on relative C turnover rates in macro- versus microaggregates or on belowground vegetation C stocks and decomposition in the rhizosphere, which are essential to understanding soil C allocation and ux. The proportion of C stocks below the active decomposition zone is important to long-term storage. What processes control the downward movement of C? The production and fate of dissolved organic C (DOC) is poorly understood, as is its potential for leaching and deep storage. Recent discoveries of large and pervasive subsurface microbial ecosystems (Fredrickson and Onstott, 1995) suggest that biogeochemical mechanisms may exist at the deep root/subsoil interface to control the fate of C at depth. Research in this area is warranted. Also in need of better understanding are the mechanisms responsible for cycling and allocation of soil C, including detrital chemistry and SOM formation, and formation and stabilization of soil micro- and macroaggregates and their role in sequestration. Soils in which it is particularly important to focus new efforts are those from tropical ecosystems, in which both soil C losses and gains are rapid at high temperature. Equally in need of additional research are: (1) C pools in frozen soils and the potential for changes in C dynamics, (2) the magnitude and dynamics of soil inorganic C (SIC), in both arid and non-arid regions, and (3) C turnover and sequestration in subsoils. Of particular importance is the need for better understanding on regional and global scales of interactions among C, water, and other major biogeochemical cycles (N, S, P, Fe) and how human activities impact these cycles. Quantitative data on relative C turnover rates in macro- versus microaggregates are limited (Jastrow et al., 1996). Recent reports on the physical fractionation of SOM have brought a new dimension to the study of SOM turnover (Cambardella and Elliott, 1993, Gregorich and Ellert, 1993). Physical separates are concrete and tangible entities. Thus, the dynamics of C and N within these fractions can be studied approximating at least the physical location where these transformations take place. Measurements of these physical entities (aggregates), however, must be accompanied by characterization of their turnover rates. Associating these to physical structures will yield viable substitutes for the current kinetic compartments of simulation models (Christensen, 1996).

28

F. BLAINE METTING ET AL.

Pedospheric processes important to regional and global C cycling include leaching, erosion, and gaseous uxes. The fate of C redistributed over the landscape by erosion and through the prole by leaching is poorly understood. One example of the possible consequences of this lack of understanding is the inability to predict the scale of potential impacts on pedospheric processes of fertilizer use and intensive agriculture in regions not previously subjected to large-scale farming, such as tropical savannahs. Research is needed to bound the probable time scales for pedospheric changes and their probable effects on C sequestration. The inuences of ongoing and expected global climate change on soil processes include increased temperature, changing moisture patterns, and increased atmospheric CO2 . Questions include: What processes govern rates of change in soil salinity and alkalinity on time scales from months to decades? What key activities of soil microbiota, microfauna, and mesofauna are central to C sequestration in months, years, and decades? How are these reected in soil productivity and quality? How do stocks and the quality of C evolve in different soils over years and decades, and how are they inuenced by initial structural, textural, and mineralogical properties? How will the susceptibility of soil to erosive forces change with addition or loss of C? What will be the time scale of these changes years or decades? 3.4. BIOTECHNOLOGY AND BIODIVERSITY Discovery and characterization efforts to understand plant and microbial diversity could be exploited for purposes of soil C sequestration. Lal et al. (1999), for example, examine in detail the research needs for discovery and application of native plant biodiversity for C sequestration and reclamation of degraded lands. Research in microbial diversity and application is currently limited because most microorganisms are unculturable. Direct extraction and study of DNA and RNA from the environment and the rapid growth in available whole microbial genome sequences are together opening new approaches to understanding and applying microbial biology. Nonetheless, the improved ability to culture important microorganisms, such as mycorrhizal fungi, remains key to advancing microbial applications to enhanced soil C sequestration. As mentioned, many or most of the properties of plants that are targets for enhanced sequestration are multigenic in nature, involving numerous metabolic and developmental molecular signaling events and pathways. Building on genome sequencing, research in functional genomics and proteomics (protein diversity and function) is needed to unravel the complexities of multigenic traits and their interaction with the environment before directed modication can proceed effectively. Thus, the use of more sophisticated molecular and instrumental approaches is important to direct basic research at key biological and ecological processes, such

INCREASING SOIL CARBON SEQUESTRATION

29

as plant-microbe molecular signaling, relatedness of environmental stress to plant NPP, and biological N2 -xation, to name but three examples. 3.5. METHODS AND INSTRUMENTATION Much of the needed research, such as improving soil C inventories, can be addressed by broader use of currently available methods. In other cases, new approaches to integrated eld and laboratory research and better instrumentation are required. The parallel improvement of process-oriented computational models for simulating key mechanisms at the appropriate molecular, pore, and meter scales, and relating them to eld and ecosystem scales is also important for understanding and predicting the potential consequences to soil C sequestration of shifting land use and global climate change (Paustian et al., 1995). Experimental approaches for addressing eld research needs include: Analog studies with soil chronosequences. Laboratory studies of mechanistic processes. Improvement of process models for describing changes in soil properties as a function of temperature, precipitation, and land use. Field manipulation studies. Long-term eld studies. Priority ecosystems for eld research to understand potential impacts of global climate change include transitional regions, coastal areas, irrigated landscapes in semiarid regions, tundra, deforested areas, and natural wetlands. Studies to understand the possible impacts of long-term trade-offs between enhanced C sequestration and reduction in gaseous C emissions from soil is needed to address the question of how different ecosystems will shift or change (Houghton et al., 1998). For example: Minimum tillage slows the rate of decomposition of SOC yet adds organic materials to the surface. Is occasional cultivation to bury the additional organic matter required to maximize long-term C sequestration? Or will this stimulate decomposition? What will be the effects of expanded use of N and other nutrient fertilization, such as in forests? Productivity will increase, but will SOM increase or will enhanced microbial activity lead to release of C to the atmosphere? Finally, new analytical methods and instruments (e.g., NMR and mass spectrometers) currently under development should be employed to improve and standardize data on soil C. Analytical methods must be standardized and widely adopted for accurate and precise quantication of different C pools while reproducibly and signicantly detecting the very small soil C concentration changes that are important on regional and global spatial scales over years, decades, and longer. Advanced isotopic and spectroscopic methods should be improved upon and used

30

F. BLAINE METTING ET AL.

on a larger scale. Development of reliable instruments for routine measurement of SOC and SIC in the eld is also an important goal. 3.6. SOCIOECONOMIC CONSIDERATIONS Research is needed to understand the overall social and economic advantages and challenges of applying new technology to enhance soil C sequestration. Central to this is the requirement for complete life cycle inventory and environmental analysis for competing land use options and for the introduction of new technologies for soil C sequestration.

4. Summary Storage of C in soils and plants has the potential to offset CO2 emissions to the atmosphere in the coming decades while new clean energy production and CO2 sequestration technologies are developed and deployed. Two-thirds of the C in the terrestrial biosphere is stored belowground. Its worldwide distribution as SOM and SIC among ecosystems is not accurately known. Also poorly understood are the biogeochemical and pedogenic mechanisms responsible for C allocation into pools of varying longevity and how the key processes are manifested at the molecular, microscopic, soil aggregate, eld, landscape, regional and global scales. Because they are economically important, have a rich history of directed research and can be most easily managed, forests and croplands are best suited for application of existing and new technology to enhance terrestrial C sequestration in the near term. Nonetheless, estimates of the potential for enhanced C storage, even in the United States, vary more than two-fold. In addition to proven management approaches, new management, chemical, and biological technology hve the potential to impact soil C storage. What is needed is basic research to improve our fundamental understanding of natural phenomena controlling soil C sequestration and basic and applied research and development to bring new management and technology to the challenge.

Acknowledgements The authors thank Norman J. Rosenberg, Rattan Lal, Steve Hildebrand, Gregg Marland, Dave Shriner and Richard Stuckey for commissioning the work that resulted in this paper and for organizing the workshop at which an earlier version was rst presented. We also thank William Schlesinger, Cynthia Cambardella, Mr. Rosenberg and Elizabeth Malone for commenting and reviewing on the original manuscript. This paper was supported, in part, under DOE contract DEAC06-76RLO 1830 to Battelle Memorial Institute to operate Pacic Northwest

INCREASING SOIL CARBON SEQUESTRATION

31

National Laboratory and former contract DE-AC05-96OR22464 to Lockheed Martin Energy Research Corporation to operate Oak Ridge National Laboratory.

Notes
1 = g dry matter per megajoule absorbed photosynthetically active radiation. 2 Thirteen states account for over 75% of the acres in CRP, a $2 billion conservation program. 3 Any scheme of this nature, however, would have to carefully consider the potential loss of CO 2

from carbonate dissolution depending on site-specic parent materials and geochemical conditions. 4 CAM = crassulacean acid metabolism. Like C-4 species, CAM plants have specialized leaf anatomy and biochemistry for CO2 xation.

References
Amthor, J. S.: 1995, Terrestrial Higher-Plant Response to Increasing Atmospheric [CO2 ] in Relation to the Global Carbon Cycle, Global Change Biol. 1, 243274. Amthor, J. S. and Huston, M. A. (eds.): 1998, Terrestrial Ecosystem Responses to Global Change. A Research Strategy, Oak Ridge National Laboratory, ORNL/TM-1998/27, Oak Ridge, TN. ap Rees, T.: 1995, Prospects of Manipulating Plant Metabolism, Trends Biotechnol. 13, 375387. Armentano, T. V. and Menges, E. S.: 1986, Patterns of Change in the Carbon Balance of Organic Soil-Wetlands of the Temperate Zone, J. Ecol. 74, 755774. Balesdent, J. and Balabane, M.: 1996, Major Contribution of Roots to Soil Carbon Storage Inferred from Maize Cultivated Soils, Soil Biol. Biochem. 28, 12611263. Batjes, N. H.: 1999, Management Options for Reducing CO2 -Concentrations in the Atmosphere by Increasing Sequestration in the Soil, Technical Paper No. 410 200 031, Dutch National Research Program on Global Air Pollution and Climate Change, Wageningen. Birdsey, R. A.: 1992, Carbon Storage in Trees and Forests, in Sampson, R. N. and Hair, D. (eds.), Forests and Global Change: Vol. 2, American Forests, Washington, D.C., pp. 2333. Bohnert, H. J. and Jensen, R. G.: 1996, Strategies for Engineering Water-Stress Tolerance in Plants, Trends Biotechnol. 14, 8991. Burke, I. C., Lauenroth, W. K., and Milchunas, D. G.: 1997, Biogeochemistry of Managed Grasslands in Central North America, in Paul, E. A. et al. (eds.), Soil Organic Matter in Temperate Agroecosystems, CRC Press, Boca Raton, pp. 85102. Cambardella, C. A. and Elliott, E. T.: 1993, Carbon and Nitrogen Distribution in Aggregates from Cultivated and Native Grassland Soils, Soil Sci. Soc. Amer. J. 57, 10711076. Christensen, B. T.: 1996, Matching Measurable Soil Organic Matter Fractions with Conceptual Pools in Simulation Models of Carbon Turnover: Revision of Model Structure, in Powlson, D. S., Smith, P., and Smith, J. U. (eds.), Evaluation of Soil Organic Matter Models, NATO ASI Series 1, Vol 38, Springer-Verlag, Berlin, pp. 143160. Cole, V., Cerri, C., Minami, K., Mosier, A., Rosenberg, N., and Sauerbeck, D.: 1996, Agricultural Options for Mitigation of Greenhouse Gas Emissions, in Watson, R. T., Zinyowera, M. C., and Moss, R. H. (eds.), Climate Change 1995: Impacts, Adaptations, and Mitigation of Climate Change: Scientic-Technical Analyses, Contribution of Working Group II to the Second Assessment Report of the Intergovernmental Panel on Climate Change, Cambridge University Press, Cambridge, U.K., pp. 744771. Coupland, R. T.: 1992, Mixed Prairie, in Coupland, R. T. (ed.), Natural Grasslands: Introduction and Western Hemisphere, Elsevier, New York, pp. 151182.

32

F. BLAINE METTING ET AL.

Department of Agriculture: 1990, Agriculture Resources: Agricultural Land Values and Markets: Situation and Outlook Report, AT-18, June 1990, Washington, D.C. Department of Energy: 1997, Technology Opportunities to Reduce U.S. Greenhouse Gas Emissions, Report by the National Laboratory Directors, http://www.ornl.gov/climate_change. Department of Energy: 1999, Carbon Sequestration Research and Development, Report by the Ofce of Science and Ofce of Fossil Energy, www.ornl.gov/carbon_sequestration. Eswaran, H., Van den Berg, E., Reich, P., and Kimble, J.: 1995, Global Soil Carbon Resources, in Lal, R., Kimble, J., Levine, E., and Stewart, B. A. (eds.), Soils and Global Change, CRC Lewis Publishers, Boca Raton. Evans, L. T.: 1993, Crop Evolution, Adaptation and Yield, Cambridge University Press, Cambridge, U.K. Fredrickson, J. K., and Onstott, T. C.: 1995, Microbes Deep Inside the Earth, Scient. Amer. 275, 6873. Furbank, R. T. and Taylor, W. C.: 1995, Regulation of Photosynthesis in C-3 and C-4 Plants: A Molecular Approach, Plant Cell 7, 797807. Gregorich, E. G. and Ellert, B. H.: 1993, Light Fraction and Macroorganic Matter in Mineral Soils, in Carter, M. R. (ed.), Soil Sampling and Methods of Analysis, Canadian Society of Soil Science, Lewis Publishers, pp. 397407. Grossman, R. B., Ahrens, R. J., Gile, L. H., Montoya, C. E., and Charwick, O. A.: 1995, Areal Evaluation of Organic and Carbonate Carbon in a Desert Area of Southern New Mexico, in Lal, R., Kimble, J., Levine, E., and Stewart, B. A. (eds.), Soils and Global Change, CRC Lewis Publishers, Boca Raton, pp. 8191. Hair, D., Sampson, R. N., and Hamilton, T. E.: 1996, Summary: Forest Management Opportunities for Increasing Carbon Storage, in Sampson, R. N. and Hair, D. (eds.), Forests and Global Change: Vol. 2, Forest Management Opportunities for Mitigating Carbon Emissions, American Forests, Washington D.C., pp. 237254. Houghton, J. T., Meira Filho, L. G., Callendar, B. A., Harris, N., Kattenberg, A., and Maskell, K. (eds.): 1996, Climate Change 1995: The Science of Climate Change, Report of the Intergovernmental Panel on Climate Change, Cambridge University Press, Cambridge, U.K. Houghton, R. A.: 1995, Changes in the Storage of Terrestrial Carbon Since 1850, in Lal, R., Kimble, J., Levine, E., and Stewart, B. A. (eds.), Soils and Global Change, CRC Lewis Publishers, Boca Raton, pp. 4565. Houghton, R. A., Davidson, E. A., and Woodwell, G. M.: 1998, Missing Sinks, Feedbacks, and Understanding the Role of Terrestrial Ecosystems in the Global Carbon Balance, Global Biogeochem. Cycles 12, 2534. Jackson, R. B., Money, H. A., and Schulze, E. D.: 1997, A Global Budget for Fine Root Biomass, Surface Area, and Nutrient Contents, Proc. Nat. Acad. Sci. 94, 73627366. Jastrow, J. D., Boutton, T. W., and Miller, R. M.: 1996, Carbon Dynamics of Aggregates-Associated Organic Matter Estimated by Carbon-13 Natural Abundance, Soil Sci. Soc. Amer. J. 60, 801807. Knutsen, G. and Metting, F. B.: 1991, Microalgal Mass Culture and Forced Development of Biological Crusts in Arid Lands, in Skujins, J. (ed.), Semiarid Lands and Deserts. Soil Resource and Reclamation, Marcel Dekker, New York, pp. 487506. Ku, M. S. B., Sakae, A., Nomura, M., Hirose, S., Toki, S., and Miyao-Tokutomi, M.: 1998, High Level Expression of Maize Phosphoenolpyruvate Carboxylase in Transgenic Rice Plants, Nature Biotechnol. 17, 7680. Kucera, C. L.: 1992, Tall-Grass Prairie, in Coupland, R. T. (ed.), Natural Grasslands: Introduction and Western Hemisphere, Elsevier, New York, pp. 227268. Lal, R., Hassan, H. M., and Dumanski, J.: 1999, Desertication Control to Sequester C and Mitigate the Greenhouse Effect, in Rosenberg, N. J., Izaurralde, R. C., and Malone, E. L. (eds.), Carbon Sequestration in Soils: Science, Monitoring and Beyond, Battelle Press, Columbus, OH.

INCREASING SOIL CARBON SEQUESTRATION

33

Lal, R., Kimble, J., and Follett, R.: 1998a, Land Use and Soil C Pools in Terrestrial Ecosystems, in Lal, R., Kimble, J., Follett, R. F., and Stewart, B. A. (eds.), Management of Carbon Sequestration in Soil, CRC Lewis Publishers, Boca Raton. Lal, R., Kimble, J., Cole, V. C., and Follett, R.: 1998b, The Potential of U.S. Cropland to Sequester Carbon and Mitigate the Greenhouse Effect, Ann Arbor Press, Ann Arbor. Lal, R., Kimble, J., and Follett, R.: 1998c, Need for Research and Need for Action, in Lal, R., Kimble, J., Follett, R. F., and Stewart, B. A. (eds.), Management of Carbon Sequestration in Soil, CRC Lewis Publishers, Boca Raton. Lauenroth, W. K. and Milchunas, D. G.: 1992, Short-Grass Steppe, in Coupland, R. T. (ed.), Natural Grasslands: Introduction and Western Hemisphere, Elsevier, New York, pp. 183226. Lloyd, J., and Farquhar, G. D.: 1994, 13 C Discrimination during CO2 Assimilation by the Terrestrial Biosphere, Oecologia 99, 201215. Metting, F. B. (ed.): 1992, Soil Microbial Ecology: Applications in Agriculture and Environmental Management, Marcel Dekker, New York. Metting, F. B.: 1996, Biodiversity and Application of Microalgae, J. Industrial Microbiol. 17, 477 489. Mummey, D. L., Smith, J. L., and Bluhm, G.: 1999, Model Estimation of Nitrous Oxide Emissions from U.S. Grasslands, Environ. Manage., in press. Oechel, W. C., Hastings, S. J., Jenkins, M., Reichers, G., Grulke, N., and Vourlitis, G.: 1993, Recent Change of Arctic Tundra Ecosystems from a Net Carbon Sink to a Source, Nature 361, 520526. Oldeman, L. R., Hakkeling, R. T. A., and Sombroek, W. G.: 1991, World Map of the Status of HumanInduced Soil Degradation: An Explanatory Note (Rev. Ed.), UNEP and ISRC, Wageningen. Parks, P. J.: 1992, Converting Marginal Crop and Pasture Land to Trees and Forest, in Sampson, R. N. and Hair, D. (eds.), Forests and Global Change: Vol. 1, Opportunities for Increasing Forest Cover, American Forests, Washington D.C. Paustian, K. H., Elliott, E. T., Collins, H. P., Cole, C. V., and Paul, E. A.: 1995, Use of a Network of Long-Term Experiments for Analysis of Soil Carbon Dynamics and Global Change: The North American Model, Aust. J. Exper. Agric. 35, 929939. Pierce, F. J. and Sadler, E. J. (eds.): 1997, The State of Site-Specic Management for Agriculture, American Society for Agronomy, Madison, WI. Ruimy, A., Saugier, B., and Dedieu, G.: 1994, Methodology for the Estimation of Terrestrial Net Primary Production from Remotely Sensed Data, J. Geophys. Res. 99, 52635283. Sage, R. F. and Monson, R. K.: 1998, C4 Plant Biology, Academic Press, San Diego. Schlesinger, W. H.: 1995, An Overview of the Carbon Cycle, in Lal, R., Kimble, J., Levine, E., and Stewart, B. A. (eds.), Soils and Global Change, CRC Lewis Publishers, Boca Raton, pp. 925. Savka, M. A. and Farrand, S. F.: 1997, Modication of Rhizobacteria by Engineering Bacterial Utilization of a Novel Plant-Produced Resource, Nature Biotechnol. 15, 363368. Schiefelbein, J. W., Masucci, J. D., and Wang, H.: 1997, Building a Root: the Control of Patterning and Morphogenesis during Root Development, Plant Cell 9, 10891098. Sewalt, V. J. H., Weiting, N., Blount, J. W., Jung, H. G., Masoud, S. A., Howles, P. A., Lamb, C., and Dixon, R. A.: 1997, Reduced Lignin Content and Altered Lignin Composition in Transgenic Tobacco Down-Regulated in Expression of L-Phenylalanine Ammonia-Lyase or Cinnamate 4Hydroxylase, Plant Physiol. 115, 4150. Stallard, R. F.: 1998, Terrestrial Sedimentation and the Carbon Cycle: Coupling Weathering and Erosion to Carbon Burial, Global Biogeochem. Cycles 12, 231257. Swinnen, J., van Veen, J. A. and Merckx, R.: 1995, Carbon Fluxes in the Rhizosphere of Winter Wheat and Spring Barley with Conventional vs. Integrated Farming, Soil Biol. Biochem. 27, 811820. Trexler, M.: 1998, Forestry as a Climate Change Mitigation Option, Appendix K in Hertzog, H. J. (ed.), Proceedings of the Stakeholders Workshop on Carbon Sequestration, Massachusetts Institute of Technology, Energy Laboratory Report No. MIT-EL 98-002.

34

F. BLAINE METTING ET AL.

Vasievich, J. M. and Alig, R. J.: 1996, Opportunities to Increase Forest Growth and Carbon Storage on Timberlands in the Contiguous United States, in Sampson, R. N. and Hair, D. (eds.), Forests and Global Change: Vol. 2, Forest Management Opportunities for Mitigating Carbon Emissions, American Forests, Washington D.C. Watson, R. T., Zinyowera, M. C., and Moss, R. H.: 1996, Climate Change 1995: Impacts, Adaptations and Mitigation of Climate Change: Scientic-Technical Analyses, Contribution of Working Group II to the Second Assessment Report of the Intergovernmental Panel on Climate Change, Cambridge University Press, Cambridge, U.K. Wright L. L., Graham, R. L., Turhollow, A. F., and English, B. C.: 1992, Growing Short-Rotation Woody Crops for Energy Production, in Sampson, R. N. and Hair, D. (eds.), Forests and Global Change: Vol. 1, Opportunities for Increasing Forest Cover, American Forests, Washington D.C., pp. 123156. Xu, J. G. and Juma, N. G.: 1992, Above- and Below-Ground Net Primary Production of Four Barley (Hordeum vulgare L.) Cultivars in Western Canada, Can. J. Plant Sci. 72, 11311140. Young, E. C. and Osborn, T. C.: 1990, Costs and Benets of the Conservation Reserve Program, J. Soil Water Conserv. 45, 370373. (Received 24 March 2000; in revised form 10 January 2001)